Exploring Musical Properties - Version 2

Chromagram

Chromagrams for own tracks

!!! I currently keep having issues with my chromagrams not loading (though they do show in R), so it may be that they don’t show up yet. Trying to fix this ASAP.

Chromagrams capture the harmonic content by showing how energy is distributed across the 12 pitch classes over time.

null device 
          1 
This took quite a lot of time to display this graphic, you may set 'fastdisp=TRUE' for a faster, but less accurate, display

Chromagrams

  • Evan-l-1.wav - Ambient
    • The energy is mainly in the lower frequency range, below ~5 kHz
    • The overall amplitude is low, this reflects the quiet ambient sounds
    • The minimal structure suggests an ambient or instrumental track
  • Evan-l-2.wav - Breakbeat
    • Covers a much wider frequency range, reaching beyond 10 kHz, with distinct variations in energy
    • Brighter colors indicate stronger amplitude, due to the track being comparatively dynamic
    • The structure implies prominent rhythmic and melodic elements, as is the case with my breakbeat track

Self-Similarity Matrices

SSMs

  • Evan-l-1.wav - Ambient
    • The chroma-SSM shows some variation, but overall pretty uniform and repetitive
    • The timbre-SSM is mostly smooth and table
  • Evan-l-2.wav - Breakbeat
    • The chroma-SSM has more distinct patterns, like a checkerboard. The song has stronger harmonic changes with varied chord progressions.
    • The timbre-SSM shows greater shifts in instruments or timbres

My AI-Generated Tracks

These tracks were created using Stableaudio AI (Stableaudio). I took inspiration from genre tags on RateYourMusic and carefully crafted prompts using detailed descriptors to shape the sound. After generating the tracks, I simply downloaded the MP3 files.

Track 1: Meditative Ambient Soundscape

Track 1: Meditative Ambient Soundscape

Style: Ambient, Post-Rock, Cinematic
Length: 2 minutes
Goal: A calm, meditative ambient with minimal instrumentation.

Tags Used:
Ambient, Post-Rock, Cinematic, Ethereal, Soothing, Meditative, Minimalist, Warm Subtle Bass, Deep Drones, Airy Pads, Textures, Analog Synths, Field Recordings, Wind Sounds, Reverb, 60 BPM

Track 2: Energetic Breakbeat Rave

Track 2: Energetic Breakbeat Rave

Style: Breakbeat, Acid Breaks, 90s Rave
Length: 2 minutes
Goal: A high-energy, chaotic breakbeat track.

Tags Used:
Breakbeat, Acid Breaks, 90s Rave, Energetic, Raw, Funky, Chaotic, Breakbeats, Deep Bass, Distorted 808, Acid Bass, Filtered Chords, Reversed Pads, Vocal Chops, 135 BPM

Creation Process

  1. Prompt Design:
    • Used RateYourMusic to explore fitting genre tags.
    • Structured prompts with specific instruments, moods, and BPM to guide the AI.
  2. Generation:
    • Entered prompts into Stableaudio AI.
    • Experimented with variations before selecting the best versions.
  3. Finalization:
    • Downloaded the MP3 files and ensured they matched my vision.

Visualization

Here’s a scatterplot of the Danceability compared to the Tempo of the tracks. My track 1 (ambient) is marked red, track 2 (breakbeat) is blue.

Final Thoughts

Findings and Final Thoughts

There appears to be no set correlation between the danceability and tempo of the tracks. However, an interesting pattern emerges: there are two clusters—one with low danceability, and another with high danceability, while the tempo does not differ much.

Regarding my own tracks:

  • The calm ambient track has an average tempo and low danceability.
  • The breakbeat song has an average tempo but high danceability.

One particularly surprising observation is how the AI interpreted the second song’s tempo. While I set it to 135 BPM, it was classified as 93 BPM. This suggests that the AI might have emphasized a different rhythmic structure or half-time feel in its classification.